. . . .
. . . .
External prolonged electrocardiogram
monitoring in unexplained syncope and
palpitations: results of the SYNARR-Flash study
E.T. Locati
1*
, A. Moya
2,3, M. Oliveira
4, H. Tanner
5, R. Willems
6,
M. Lunati
1, and M. Brignole
71
Department of Cardiovascular—Cardiology 3, Electrophysiology, Ospedale Niguarda Ca’ Granda, Piazza Ospedale Maggiore, 3, Milano 20162, Italy;2
Univeristy Hospital Vall d’Hebron, Barcelona, Spain;3
University Hospital QuironDexeus, Barcelona, Spain;4
Hospital Santa Marta, Lisbon, Portugal;5
Inselspital, Bern, Switzerland;6
Uz Hospital, Leuven, Belgium; and7
Lavagna Hospital, Lavagna, Genova, Italy
Received 14 July 2015; accepted after revision 18 August 2015; online publish-ahead-of-print 29 October 2015
Aims SYNARR-Flash study (Monitoring of SYNcopes and/or sustained palpitations of suspected ARRhythmic origin) is an
inter-national, multicentre, observational, prospective trial designed to evaluate the role of external 4-week electrocardiogram (ECG) monitoring in clinical work-up of unexplained syncope and/or sustained palpitations of suspected arrhythmic origin.
Methods and results
Consecutive patients were enrolled within 1 month after unexplained syncope or palpitations (index event) after being discharged from emergency room or hospitalization without a conclusive diagnosis. A 4-week ECG monitoring was
obtained by external high-capacity loop recorder (SpiderFlash-Tw
, Sorin) storing patient-activated and auto-triggered tracings. Diagnostic monitorings included (i) conclusive events with reoccurrence of syncope or palpitation with con-comitant ECG recording (with/without arrhythmias) and (ii) events with asymptomatic predefined significant arrhyth-mias (sustained supraventricular or ventricular tachycardia, advanced atrio-ventricular block, sinus bradycardia ,30 b.p.m., pauses .6 s). SYNARR-Flash study enrolled 395 patients (57.7% females, 56.9 + 18.7 years, 28.1% with syncope, and 71.9% with palpitations) from 10 European centres. For syncope, the 4-week diagnostic yield was 24.5%, and predictors of diagnostic events were early start of recording (0 – 15 vs. .15 days after index event) (OR 6.2, 95% CI 1.3 – 29.6, P ¼ 0.021) and previous history of supraventricular arrhythmias (OR 3.6, 95% CI 1.4 – 9.7, P ¼ 0.018). For palpitations, the 4-week diagnostic yield was 71.6% and predictors of diagnostic events were history of recurrent palpitations (P , 0.001) and early start of recording (P ¼ 0.001).
Conclusion The 4-week external ECG monitoring can be considered as first-line tool in the diagnostic work-up of syncope and
palpitation. Early recorder use, history of supraventricular arrhythmia, and frequent previous events increased the likelihood of diagnostic events during the 4-week external ECG monitoring.
-Keywords External loop recorder † Syncope † Palpitation † Arrhythmia diagnosis † Ambulatory ECG monitoring
Introduction
The diagnosis of unexplained syncopal episodes and sustained palpi-tations remains a difficult task in clinical cardiology.1–6Standard 24 h Holter monitoring has a low diagnostic yield, while implantable loop recorder (ILR) has a higher diagnostic yield, but it is expensive and mildly invasive, making its role in diagnostic work-up of unexplained syncope and palpitations questionable.7–10
The utility of external prolonged electrocardiogram (ECG) moni-toring in work-up of syncope is still undefined. The few available stud-ies, generally single-centre and retrospective, provided conflicting results, mainly due to heterogeneity of patient populations and devices capabilities. In sustained palpitations, the external loop recorders (ELRs) utility was more established.5Earlier ELRs with relatively low-storage capacity and brief patient-activated ECG tracings had limited usefulness in syncope or asymptomatic arrhythmias.1,7,11–14Recently,
*Corresponding author. Tel:+39 02 6444 2069; fax: +39 02 6444 2566. E-mail address: emanuelateresa.locati@ospedaleniguarda.it
&The Author 2015. Published by Oxford University Press on behalf of the European Society of Cardiology.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact
newer ELRs with auto-trigger capabilities showed 30% diagnostic
yield for syncope and 75% for palpitations.15External ECG recorders
were also utilized to evaluate the burden of asymptomatic atrial fibrillation (AF), as in cryptogenic transient ischaemic attacks or after transcatheter ablation.16,17
Syncope and palpitations tend to occur in clusters, with higher re-occurrence rate early after an event, thus an early use of long-term ECG monitoring may increase the likelihood of clinical diagnosis.1,10 The SYNARR-Flash study (Monitoring of SYNcopes and/or sus-tained palpitations of suspected ARRhythmic origin; clinicaltrial.gov NCT02253134) is the first international, multicentre, prospective trial designed to evaluate the feasibility and usefulness of external pro-longed ECG monitoring in the early clinical work-up of unexplained syncope and/or sustained palpitations of suspected arrhythmic origin.
Methods
Study design
At each enrolling centre, consecutive patients screened for enrolment had to meet both the following inclusion criteria: (i) recent (within 1 month) episode of syncope or sustained palpitations (index event), after being discharged from emergency room or hospitalization without a conclusive diagnosis, and (ii) suspected arrhythmic origin according to the clinical features defined in the 2009 Syncope Guidelines, including the presence of cardiovascular disease or channelopathy, family history of unexplained sudden death, or abnormal ECG findings.1
Syncope was defined as complete loss of consciousness with spontan-eous recovery, while sustained palpitations as long-lasting sensation of irregular or fast heart rhythm. Abnormal ECG findings included atrio-ventricular conduction disorders [first-degree atrio-atrio-ventricular block (AVB) I, second-degree AVB II type Mobitz 1, left or right ventricular bun-dle branch block (LBBB/RBBB)], supraventricular rhythm disorders [sinus arrest or pauses ,3 s, sinus bradycardia .40 b.p.m., sinus tachycardia .100 b.p.m., paroxysmal AF, frequent premature atrial contractions (PACs) (.10/min), chronic AF, paroxysmal atrial flutter/tachycardia], and ventricular rhythm disorders [frequent premature ventricular con-tractions (PVCs) (.10/min), couplets, non-sustained VT] (Table1).
By study inclusion criteria, all patients were enrolled within 30 days after the index event, after being discharged from emergency room or hospitalization without a conclusive diagnosis. At enrolment, patient medical history was collected, including a detailed report of the index event (syncope or palpitation) and the history of events in the preceding year and of the previously performed diagnostic tests.
Patients were instructed by the study nurse on how to use the re-corder and to report symptoms in the diary. Patients had to wear the recorder for 4 weeks, and a clinic visit was scheduled at the end of the recording period (M1). Patients had to contact the centre if any symptom occurred before M1 (in case an additional clinic visit was scheduled) or if any technical problem occurred or the recording stopped prematurely (in order to re-initiate the recording).
At M1, patients were interviewed, diaries were collected, and infor-mation was cross-checked. Electrocardiogram data were uploaded to the server. Clinical and device information, including events, device acceptance, device malfunction, adverse events, and medications, were reported in case report forms (CRFs).
The study was conducted in accordance with Guidelines for Good Clinical Practice and Declaration of Helsinki and approved by relevant local ethics committees. All enrolled patients provided written informed consent to study participation.
Modality of ECG recording
At enrolment, a prolonged high-capacity ELR (SpiderFlash-Tw
, Sorin Group SRL, Saluggia, Italy—sized 75× 50 × 19 mm, powered by lith-ium battery) was provided to each patient. The three-wire recorder was connected to the chest using disposable adhesive electrodes, which the patient had to change daily. The recorder was carried in a disposable bag placed around the neck.
Recorders store two leads ECG tracings on a high-capacity removable digital card capable of up to 40 day monitoring. Three recording modal-ities are available: (i) manual patient activation in case of symptoms; (ii) automatic activation at predefined intervals; and (iii) auto-trigger activa-tion at preselected rhythm disturbances, such as pauses or bradycardia or supraventricular tachycardia (SVT) and ventricular tachycardia (VT).
Event classification and computation
of diagnostic yield
All ECG monitorings were reviewed by two separate investigators (E.T.L. and A.M.) and categorized based on both patient diary and inter-view and ECG data, as follows.
Diagnostic monitorings include the following:
† Conclusive events, when syncope or palpitations were reported and a readable ECG recording was available at the time of symptom (inde-pendently of the presence or absence of arrhythmias);
† Significant events, when the following predefined arrhythmias were detected by auto-trigger function in the absence of reported symptoms: (i) advanced AVB (third-degree AVB or second-degree AVB Mobitz type 2), sinus bradycardia (,30 b.p.m.), pauses≥6 s; (ii) fast sustained supra-ventricular tachycardia (SVT), AF, or flutter (AFL) (rate .180 b.p.m., duration .3 min); and (iii) non-sustained ventricular tachycardia (NSVT, duration .10 s) or sustained ventricular tachycardia (.30 s). Those arrhythmias were considered diagnostic even in the absence of recurrent syncope according to the 2009 Syncope Guidelines.1 Non-diagnostic monitorings include the following:
† Suggestive events, when the following predefined arrhythmias were detected by auto-trigger function, in the absence of reported symp-toms: (i) sinus bradycardia (30 – 40 b.p.m.), 3 – 6 s pauses; (ii) brief burst of SVT, AF, or AFL (15 – 180 s); and (iii) NSVT ,10 s;
What’s new?
† The early use of external 4-week electrocardiogram (ECG) recorders in unexplained syncope of suspected arrhythmic origin proved to be a feasible ‘stepwise’ strategy, starting the recording as soon as possible after an event in order to maximize the diagnostic yield of external ECG monitoring. † In patients with unexplained syncope of suspected arrhythmic
origin, in whom prolonged ECG monitoring is considered ap-propriate according to current guidelines, the 4-week external ECG monitoring should be utilized as first step, while more ex-pensive and minimally invasive implantable loop recorder (ILR) should be reserved to those cases who remained undiagnosed after the 4-week external monitoring.
† In patients with unexplained palpitation, the 4-week external ECG monitoring can be considered as first-line diagnostic tool, providing a conclusive diagnosis in most cases, avoiding use-less repetition of standard 24 h Holter monitoring, with longer monitoring by ILR would be required only in few selected cases.
† Negative monitorings: (i) monitorings without event recurrence reported in the patient diary and without recording of asymptomatic arrhythmia by auto-trigger function; (ii) monitorings with recurrence of symptoms different from the index event; and (iii) monitorings with recurrence of syncope or palpitations, but without available ECG recording.
Statistical analysis
All analyses were performed on the intention-to-treat (ITT) popu-lation, both overall and separately per index event (syncope or palpitations). Since the distribution of the data was normal, continu-ous data were expressed as mean + standard deviation (SD), while categorical variables as frequencies and percentages and differences in categorical variables were tested by x2test or by Fisher exact test as appropriate.
The time-to-first occurrence of a diagnostic event during 4-week monitoring was analysed by Kaplan – Meier survival analysis, and sur-vival curves of patients with syncope and palpitations were com-pared by log-rank test. Actuarial diagnostic yields for diagnostic events were retrieved from Kaplan – Meier statistics and compared by x2analysis when applicable.
Predictors of occurrence of diagnostic events during the 4-week monitoring were identified by multivariable Cox
proportional-hazard model, after proportional-proportional-hazard assumptions were verified. A stepwise procedure was applied: first, univariable analyses deter-mined statistically significant variables at 20%, and then significant variables were introduced in the multivariable model.
Two-sided P values of ,0.05 were considered significant in all evaluations. Analyses were performed using SAS software release 9.2 (SAS Institute, Cary, NC, USA).
Results
Study population
The SYNARR-Flash study enrolled 395 patients (42.3% males, age 56.9 + 18.7 years) from 10 centres in five countries between August 2010 and June 2013. As index event information was unavail-able for three patients, the ITT population included 392 patients; 282 patients (71.9%) enrolled for palpitations and 110 (28.1%) for syncope.
Patients with syncope were older, more frequently males, with previous history of cardiovascular disease, family history of sudden
death, hypertension, and use of cardiovascular medication (Table1).
Previous diagnostic tests, when available, were reported negative in all patients.
. . . . Table 1 Demographic characteristics
Parameters (number, %) ITT population with syncope n 5 110
ITT population with palpitation n 5 282 P-value Demographics Age, mean (+SD) 65.1 (+17.2) 53.9 (+18.4) ,0.001 Male gender 64 (59.3%) 101 (35.8%) ,0.001 NYHA, n 68 191 I/II 66 (97.1%) 190 (99.5%) 0.170 III/IV 2 (2.9%) 1 (0.5%)
Previous diagnosis test, n 106 262
24 h Holter recording 64 (60.4%) 159 (60.7%) 0.956
HUT 21 (19.8%) 13 (5.0%) ,0.001
CSM 27 (25.5%) 10 (3.8%) ,0.001
Exercise stress test 12 (11.3%) 41 (15.6%) 0.284
History of previous rhythm disorders
Supraventricular rhythm disordersa 40 (57.1%) 103 (66.0%) 0.200 AV conduction disordersb 32 (45.7%) 63 (40.4%) 0.453 Ventricular rhythm disordersc 29 (41.4%) 65 (41.7%) 0.973 History of cardiovascular disease, n 70 155
Systemic hypertension 41 (58.6%) 68 (43.9%) 0.041
Coronary artery disease 8 (11.4%) 12 (7.7%) 0.368
Cerebrovascular accident 8 (11.4%) 7 (4.5%) 0.054
Ischaemic cardiomyopathy 5 (7.1%) 9 (5.8%) 0.701
Previous myocardial infarction 3 (4.3%) 6 (3.9%) 1.000 Family history of sudden cardiac death 9 (8.2%) 7 (2.5%) 0.006 Cardiovascular medication at the time of the test 82 (74.5%) 190 (67.4%) 0.166
SD, standard deviation; NYHA, New York Heart Association functional class; AV, atrio-ventricular.
a
Includes sinus arrest/pauses ,3 s, sinus bradycardia .40 b.p.m., sinus tachycardia, paroxysmal AF, frequent PACs (.10/min), chronic AF, and paroxysmal atrial flutter/ tachycardia.
b
Includes AVB I, AVB II Mobitz 1, LBBB, and RBBB.
c
Duration of ECG monitoring
The mean duration of ECG monitoring was 23.0 + 8.1 days, similar for syncope and palpitations. In 22 patients, the recording was stopped for the occurrence of a major event. In four patients (1%), the recording was interrupted due to intolerance to electro-des, while two patients (0.5%) withdrew from the study. Skin reac-tions to electrodes were reported in 12 patients (3%), but all were defined as non-serious and solved spontaneously.
Findings in patients with syncope
Twenty-seven patients with prior syncope (24.5%) had a diagnos-tic event within the 4-week monitoring. Conclusive events with
ECG-documented syncope occurred in 11 patients (10%, Table2):
in 6 patients, supraventricular bradyarrhythmias were recorded during syncope, while in 5 cases, regular sinus rhythm or sinus
tachycardia was recorded (Table2). Auto-trigger function
identi-fied asymptomatic significant arrhythmias in 16 patients (14.5%, Table2).
The actuarial diagnostic yield, calculated by the Kaplan – Meier cu-mulative occurrence of diagnostic events, was 13.2% at 1 week,
19.1% at 2 weeks, and 29.4% at 4 weeks (Figure1).
Auto-trigger function detected asymptomatic suggestive
arrhyth-mias in 10 patients (9.1%, Table2), although per definition were not
included as diagnostic events. The cumulative yield including con-clusive, significant, and suggestive events would be 33.6%.
Multivariable analyses to identify potential predictors of diagnos-tic events included age, gender, previous recording of supraventricu-lar or ventricusupraventricu-lar arrhythmia, time from index event to enrolment (0 – 15 days vs. .15 days after index syncope), the number of events in preceding year, history of cardiovascular disease, and family his-tory of sudden cardiac death. The only two significant predictors
. . . . . . . . Table 2 Electrocardiogram findings during monitoring in patients with syncope and palpitation
Patients Syncope (n 5 110) Palpitation (n 5 282)
n % n %
Diagnostic tests 27 24.5 202 71.6
Conclusive events (recurrence reported on patient diary) 11 10.0 193 68.4 Non-arrhythmic (sinus rhythm, sinus tachycardia) 5 4.5 66 23.4
Arrhythmic 6 5.5 127 45.0
Pauses (≥6 s), AVB 3 or AVB 2 Mobitz 2 4 3.6 1 0.4
Pauses (3 – 6 s), sinus bradycardia 2 1.8 7 2.5
Paroxysmal AF, AFL (.3 min) 0 0.0 22 7.8
Paroxysmal SVT (.3 min) 0 0.0 15 5.3
Non-sustained SVT (15 s – 3 min) 0 0.0 40 14.2
NSVT ,10 s 0 0.0 5 1.8
NSVT .10 s 0 0.0 1 0.4
Frequent VPBs or APBs 0 0.0 36 12.7
Asymptomatic significant arrhythmias 16 14.5 9 3.2
Fast sustained AF/AFL/SVT .3 min .180 b.p.m. 8 7.3 8 2.8 Sinus bradycardia ,30 b.p.m., Pauses≥6 s, AVB 3 or AVB 2 Mobitz 2 5 4.5 1 0.4 NSVT .10 s or sustained ventricular tachycardia .30 s 3 2.7 0 0.0
Non-diagnostic tests 83 75.5 80 28.4
Negative monitoring 73 66.4 59 20.9
Asymptomatic suggestive arrhythmias 10 9.1 21 7.5
Sinus bradycardia 30 – 40 b.p.m., sinus pauses 3 – 6 s 5 4.5 11 3.9
NSVT ,10 s 5 4.5 9 3.2
AF/AFL/SVT 15 s – 3 min 0 0.0 1 0.4
AVB 2, AVB 3, atrio-ventricular block (Grade 2 or 3); AF, atrial fibrillation; AFL, atrial flutter; SVT, supraventricular tachycardia; NSVT, non-sustained ventricular tachycardia; VPBs, ventricular premature beats; APBs, atrial premature beats.
0.9 0.8 Log rank : P < 0.001 0.7 0.6 F reedom from e v ent 0.5 0.4 0.3 0.2 105 88 81 48 47 14 15 4 7 157 101 264 Number of patients at risk : 0.1 0 0 7 14
ITT Syncope ITT Palpitation
Time to first diagnosed event (days)
21 28 35
1
Figure 1 Kaplan – Meier analysis of recurrence-free rate in pa-tients with syncope (continuous line) and palpitations (dashed line) of first diagnostic event during prolonged ECG monitoring (log rank: P , 0.001).
for diagnostic findings during the 4-week monitoring were (i) early start of monitoring after index event and (ii) previous history of supraventricular arrhythmias [sinus arrest or pauses ,3 s, sinus bradycardia .40 b.p.m., sinus tachycardia .100 b.p.m., paroxysmal AF, frequent PACs (.10/min), chronic AF, paroxysmal atrial flutter/
tachycardia—Table3]. Monitoring was started within 15 days from
index event in 74 patients: 4-week actuarial diagnostic yield was 33.0% in patients with early initiation compared with 15.6% in later initiation (P ¼ 0.021, Figure2A). History of supraventricular arrhyth-mias was present in 39 patients: 4-week actuarial diagnostic yield was 43.6% in patients with supraventricular arrhythmias compared
with 21.3% in those without (P ¼ 0.018, Figure2B).
Findings in patients with palpitations
A total of 202 patients (71.6%) had a diagnostic event within the
4-week monitoring (Table2). Conclusive events occurred in 193
pa-tients (68.4%). Sinus rhythm or sinus tachycardia was present at patient-activated recording during palpitation in about two-thirds of the cases, while SVT, AF or AFL, bradycardia, pauses or NSVT were present in the remaining cases. Significant asymptomatic arrhythmias were detected by auto-trigger function in nine patients (3.2%).
Asymptomatic suggestive arrhythmias were detected by
auto-trigger function in 21 patients (7.5%, Table2), which per definition
were not included as diagnostic events. The cumulative yield includ-ing conclusive, significant, and suggestive events would be 79.1%.
The Kaplan – Meier cumulative occurrence of diagnostic events was
42.4% at 1 week, 57.2% at 2 weeks, and 77.0% at 4 weeks (Figure1).
Previous frequent palpitations and early start of recording after the index event were significant predictors of diagnostic events (Table3).
The diagnostic yield progressively increased with the number of
palpitations during the year preceding the enrolment (Figure3A).
The diagnostic yield was 46.2% in patients with 1 or no palpitation, 68.1% in patients with 2 – 5 palpitations, 76.7% in patients with 6 – 10 palpitations, and 83.1% in patients with≥11 palpitations (Figure3A). The 4-week diagnostic yield was higher in patients starting the recording 0 – 7 days after the index event (Table3, Figure3B).
Discussion
The SYNARR-Flash study is the first international, multicentre, obser-vational, prospective trial designed to prove the feasibility and
. . . . Table 3 Predictors for diagnosis during the 4-week ECG monitoring
Variables P-value Odds ratio 95% Wald
confidence limits
A—patients studied for syncope
Time between index event and enrolment 0 – 15 days vs. .15 days 0.021 6.2 1.3 29.6 History of supraventricular arrhythmiasa Yes vs. No 0.018 3.6 1.4 9.7 B—patients studied for palpitations
History of previous palpitations (in the previous 12 months) 2 – 5 vs.≤1 0.036 2.6 1.1 6.3 6 – 10 vs.≤1 0.029 2.8 1.1 7.3 ≥11 vs. ≤1 ,0.001 4.3 2.0 9.4 Time between index event and enrolment 0 – 7 days vs. 8 – 15 days 0.004 3.0 1.4 6.4 0 – 7 days vs. .15 days 0.031 2.3 1.1 4.9
a
Supraventricular arrhythmias include sinus arrest/pauses ,3 s, sinus bradycardia .40 b.p.m., sinus tachycardia, paroxysmal AF, frequent PACs (.10/min), chronic AF, and paroxysmal atrial flutter/tachycardia.
1 0.9 A B 0.8 Log rank: P = 0.021 Log rank: P = 0.018 0.7 0.6 F
reedom from syncope
F
reedom from syncope
0.5 0.4 0.3 0.2 0.1 74 61 25 56 24 34 14 12 2 28 62 55 30 53 26 33 13 9 3 39 Number of patients at risk: Number of patients at risk: 0 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2
No supraventricular rhythm discorder Supraventricular rhythm discorder
0.1 0
0 7
0–15 days >15 days
14
Time to first diagnosed event (days)
21 28
0 7 14
Time to first diagnosed event (days)
21 28
Figure 2 (A) Kaplan –Meier analysis of recurrence-free rate in pa-tients with syncope enrolled at 0– 15 days after the index syncope (continuous line) vs. .15 days after the index syncope (dashed line), OR ¼ 6.222, 95% CI (1.309– 29.573), P ¼ 0.02 (see Table3for de-tails). (B) Kaplan– Meier analysis of recurrence-free rate in patients with syncope without previous history of supraventricular rhythm disorder (continuous line) vs. patients with previous history of su-praventricular rhythm disorder (dashed line), OR ¼ 3.631, 95% CI (1.356– 9.724), P ¼ 0.01 (see Table3for details).
usefulness of external prolonged ECG monitoring in early clinical work-up of unexplained syncope and/or sustained palpitations of suspected arrhythmic origin. In unexplained syncope, prolonged external ECG monitoring achieved a diagnosis in about one-third of the patients, and most diagnoses occurred in those who started
the monitoring within 15 days from index syncope (Figure2A) and
in those with history of SVT (Figure2B). These findings suggest two
new clinical criteria in order to increase the diagnostic likelihood in unexplained syncope, namely early initiation of monitoring after an event and previous history of supraventricular arrhythmias. In unexplained palpitations, the 4-week actuarial diagnostic yield was 77.0%, increasing up to 81.3% in patients with previous fre-quent palpitations (Figure3A), supporting its utility as first-line diag-nostic tool in palpitation work-up.
Prolonged external ECG monitoring
in unexplained syncope
All enrolled patients had unexplained syncope of suspected arrhyth-mic origin according to the criteria outlined in 2009 Guidelines for
diagnosis and management of syncope.1In this study, the actuarial
diagnostic yield of the 4-week ECG monitoring was 29.4%, higher than the 1 – 10% diagnostic yield generally reported for 24 h Holter recording, and consistent with previous studies utilizing external
long-term ECG monitoring, reporting diagnostic yields of 15 – 20% at 1 month.11–15,19
In our study, the 4-week diagnostic yield for syncope was similar or higher than in studies using ILR, when considering a similar
mon-itoring period. In the PICTURE registry,9the 3-month diagnostic
yield was 19%, and in most studies, the diagnostic yield of ILR ranged
from 30 up to 50%,7–10indicating that syncope often remained
un-explained even after 3-year ILR monitoring. These data support the concept that diagnostic findings in most patients with unexplained syncope occur relatively early after an index event, making pro-longed external ECG monitoring sufficient to obtain a clinical diag-nosis in most cases, with only a few selected patients requiring a much longer monitoring period.
In our study, arrhythmias were documented during syncope in about half of the patients, typically bradycardia or pauses leading to pacemaker implant, while in the remaining cases, no arrhythmias were detected, excluding an arrhythmic origin of syncope. These pa-tients should then be referred for further clinical evaluation, possibly including carotid sinus massage (CSM) or head-up tilt test (HUT) or neurological work-up, while further prolonged ECG monitoring is not necessary.
Asymptomatic arrhythmias predefined as significant were
de-tected by auto-trigger function in 16 patients (14.5%, Table2).
Sig-nificant arrhythmias were considered diagnostic even in the absence of recurrent syncope following the 2009 Syncope Guidelines, with
evidence Class 1 Level C.1This classification was specifically
pro-posed by the International Study on Syncope of Unknown Etiology (ISSUE) investigators, with the aim to group the observations from prolonged ECG monitoring into homogeneous patterns in order to define an acceptable standard useful for future studies and clinical
practice.20 Of note, this classification has been utilized in most
studies with ILR, where unsuspected asymptomatic arrhythmias in the absence of syncope were generally considered diagnostic findings.7–10
To avoid potential bias, for this study, we utilized more stringent criteria for definition of significant asymptomatic arrhythmias than those proposed by the 2009 Syncope Guidelines. Asymptomatic
suggestive arrhythmias were detected in 9.1% of patients (Table2),
which per definition were not included in the diagnostic yield, although some of these arrhythmias, such as 3 – 6 s pauses, would have been considered diagnostic according to 2009 Syncope
Guidelines.1,20Patients with asymptomatic suggestive arrhythmias
seem to be the ideal candidates to continue prolonged ECG mon-itoring by ILR, to confirm or exclude the presence of significant arrhythmias.
In syncope, early initiation of recording after an index event was a significant predictor of diagnosis during the 4-week
moni-toring (Figure2A). This is consistent with some preliminary
obser-vations suggesting that syncopal events occur in clusters,10making
it crucial to initiate the monitoring as soon as possible after an event.
A second significant predictor of diagnostic events was history of supraventricular arrhythmias, suggesting that patients with previous paroxysmal AF are reasonable candidates to prolonged ECG mon-itoring after an unexplained syncope. Of note, AF was not included among ECG criteria suggesting an arrhythmic origin for unexplained
syncope in 2009 Guidelines.1 1 0.9 A B Log rank: P < 0.001 Log rank: P < 0.001 0.8 0.7 0.6 F
reedom from palpitation
F
reedom from palpitation
0.5 0.4 0.3 Number of patients at risk : Number of patients at risk : 51 45 41 111
History of events (last 12 months) 40 28 21
159
Time between index event and enrollment: 49 48 83 40 31 52 25 21 27 9 9 10 1 7 4 0 3 56 32 12 14 36 ≤1 15 5 8 14 6 2 3 7 2 1 2 1 0.2 0.1 0 0 7 14 2 to 5 6 to 10 ≥11
Time to first diagnosed event (days)
Freedom from first event leading to diagnosis
Time to first diagnosed event (days)
21 28 35
0 7 14
0–7 days 8–15 days >15 days
21 28 35 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
Figure 3 (A) Kaplan – Meier analysis of recurrence-free rate in patients with palpitations according to the previous number of events in the last 12 months (see Table3for details). (B) Ka-plan – Meier analysis of recurrence-free rate in patients with palpi-tations according to the time between index event and enrolment (see Table3for details).
Prolonged external ECG monitoring
in unexplained palpitations
In patients with unexplained sustained palpitations, the diagnostic yield of external monitoring was 42.4% at 1 week, 57.2% at 2 weeks, and 71.6% at 4 weeks. These data are consistent with previous stud-ies,5,15,20,21making the 4-week external ECG monitoring the first-choice tool in the diagnostic workflow of unexplained palpitations. Only patients with palpitations remaining unexplained after the 4-week monitoring may require further observation, either by ILR21or by event recorders.22
The diagnostic yield was50% at 1 week and 70% at 2 weeks in
patients with previous frequent palpitations (Figure3A) or in patients studied early after an index event (Figure3B). These data support the utility of 1-week and 2-week external recordings, especially when utilized in high-risk patients and started early after an event.
Sinus rhythm or sinus tachycardia, or short episodes of SVT were observed during palpitations in most patients, while sustained and
fast SVT, AF, or AFL were documented in 13% of the cases (Table2).
In fewer cases, sinus bradycardia and pauses were observed, often associated with dizziness or presyncope, besides palpitations. The clinical significance or the clinical management of such findings was beyond the scope of this study.
Although the main objective of prolonged external ECG moni-toring was to correlate symptoms and arrhythmias, asymptomatic paroxysmal AF (either brief or sustained episodes) was detected
in10% of the patients with unexplained palpitations. These
find-ings support the discrepancy between perceived symptoms and documented arrhythmias, and confirm that silent AF is relatively frequent even in patients with history of palpitations.4,5,16,17
Study limitations
One possible limitation was that not all centres contributed with the same amounts of patients, as three centres (Valld’Hebron, Milan,
and Lavagna) enrolled75% of the patients. However, no
signifi-cant differences were observed among patients enrolled by the different centres.
The information provided in patients’ diary was used to cat-egorize for the absence or presence and kind of symptom (syn-cope or palpitation) at the time of arrhythmias. It is impossible to verify the accuracy of events reported in the diary by each pa-tient, although the diary information was cross-checked at M1 clinic visits.
Predefined criteria were utilized to distinguish between ‘significant’ and ‘suggestive’ asymptomatic arrhythmias detected by auto-trigger
function, based on recommendation from current guidelines1,20and
on our best clinical judgement. Due to lack of consensus on the
min-imum duration required to define an episode of paroxysmal AF,4,5we
utilized a restrictive criteria (duration .180 s) for significant events, while shorter events were only considered suggestive. Although the presence of fast and sustained supraventricular arrhythmias may sug-gest a syncope of arrhythmic origin, the precise mechanism provoking the syncope in the individual patient remains unknown, and the attri-bution of a specific aetiology for syncope in the individual patients is beyond the scope of this study. More stringent criteria were used also for pauses and NSVT, although some suggestive asymptomatic ar-rhythmias may represent meaningful findings in the diagnostic
workflow and would have been considered diagnostic according to the 2009 Syncope Guidelines.1,20
Information about previous diagnostic tests, specifically CSM and HUT, was collected wherever available, but it was beyond the power of this study to verify why a specific diagnostic test was per-formed or not perper-formed in the single patient. This study population was too small to draw any definitive conclusion on the possible cor-relation between ELR-negative and HUT-negative findings.
The cumulative diagnostic yield observed in this study both for syncope and palpitation may be an overestimation of the true clinical benefit. The capability of prolonged early monitoring to influence therapeutic decisions or to improve clinical outcomes was beyond of the scope of this study and remains to be demonstrated by an appropriately designed study.
Clinical implications and
conclusions
The early use of external 4-week ECG recorders in unexplained syncope of suspected arrhythmic origin proved to be a feasible ‘stepwise’ strategy, starting the recording as soon as possible after an event in order to maximize the diagnostic yield of external ECG monitoring. In patients with unexplained syncope of suspected ar-rhythmic origin, in whom prolonged ECG monitoring is consid-ered appropriate according to current guidelines, the 4-week external ECG monitoring should be utilized as first step, while more expensive and minimally invasive ILR should be reserved to those cases who remained undiagnosed after the 4-week exter-nal monitoring. In patients with unexplained palpitation, the 4-week external ECG monitoring can be considered as first-line diagnostic tool, providing a conclusive diagnosis in most cases, avoiding useless repetition of standard 24 h Holter monitoring, with longer monitoring by ILR would be required only in a few se-lected cases.22–25
The results of this study, which utilized external loop recording with auto-trigger function, may be extended to new systems of long-lasting external ECG recordings providing continuous ECG
monitoring,21that may have an even higher capability of detecting
asymptomatic arrhythmias.
In patients with unexplained palpitation, the 4-week external ECG monitoring can be considered as first-line diagnostic tool, pro-viding a conclusive diagnosis in most cases, avoiding useless repeti-tion of standard 24 h Holter monitoring, with longer monitoring by ILR required only in few selected cases.
Acknowledgements
The authors thank all the doctors and register nurses from the 10 enrolling centres who contributed to patients’ enrolment and clinic visits, and specially Dr J.M. Ormaetxe (Basurto Hospital, Bilbao), Dr M. Lopez-Gil (October 12 University Hospital, Madrid), Dr X. Sabate (Bellvitge University Hospital, Barcelona), and Dr N. Rivas (University Hospital Valld’Hebron and University Hospital Quiron Dexeus, Barcelona). The authors thank Mara Rolando, Andrea Pinciroli, Frederique Maneval, and Alessandra Pezzotta, from Sorin Group, for technical support and Franca Negrini, RN, for study assistance.
Funding
This study was supported by a research grant of SORIN GROUP, which supplied the SPIDERFLASH-T recorders free of charge to all enrolling centres and provided logistic support and assistance for database man-agement and statistical analysis. R.W. was supported as a clinical researcher by the Fund for Scientific Research Flanders (FWO). Funding to pay the Open Access publication charges for this article was provided by SORIN GROUP.
Conflict of interest: none declared.
References
1. Moya A, Sutton R, Ammirati F, Blanc JJ, Brignole M, Dahm JB et al. Guidelines for the diagnosis and management of syncope (version 2009). Eur Heart J 2009;30:2631–71. 2. Zipes DP, Camm AJ, Borggrefe M, Buxton AE, Chaitman B, Fromer M et al. ACC/ AHA/ESC 2006 guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death. Europace 2006;8:746 – 837. 3. Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S et al. Guidelines for
the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace 2010;12: 1360 – 420.
4. January CT, Wann L, Alpert JS, Calkins H, Cigarroa JE, Cleveland JC Jr et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Soci-ety. J Am CollCardiol 2014;64:2246 – 80. doi:10.1016/j.jacc.2014.03.021. 5. Raviele A, Giada F, Bergfeldt L, Blanc JJ, Blomstrom-Lundqvist C, Mont L et al.
Man-agement of patients with palpitation: a position paper from the European Heart Rhythm Association. Europace 2011;13:920 – 34.
6. Priori SG, Blomstro¨m-Lundqvist C, Mazzanti A, Blom N, Borggrefe M, Camm J et al. 2015 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death. Europace 2015;17:1601 – 87. 7. Brignole M, Vardas P, Hoffman E, Huikuri H, Moya A, Ricci R et al. Indications for the
use of diagnostic implantable and external ECG loop recorders. Europace 2009;11: 671 – 87.
8. Hong P, Sulke N. Implantable diagnostic monitors in the early assessment of syncope and collapse. Prog Cardiovasc Dis 2013;55:410 – 7.
9. Edvardsson N, Frykman V, van Mechelen R, Mitro P, Mohii-Oskarsson A, Pasquie´ JL et al. Use of an implantable loop recorder to increase the diagnostic yield in unex-plained syncope: results from the PICTURE registry. Europace 2011;13:262 – 9. 10. Brignole M, Sutton R, Menozzi C, Garcia-Civera R, Moya A, Wieling W et al. Early
application of an implantable loop recorder allows effective specific therapy in pa-tients with recurrent suspected neurally mediated syncope. Eur Heart J 2006;27: 1085 – 92.
11. Schuchert A, Maas R, Kretzschmar C, Behrens G, Kratzmann I, Meinertz T. Diag-nostic yield of external electrocardiographic loop recorders in patients with recur-rent syncope and negative tilt table test. Pacing Clin Electrophysiol 2003;26:1837 – 40. 12. Sivakumaran S, Krahn AD, Klein GJ, Finan J, Yee R, Renner S et al. A prospective randomized comparison of loop recorders versus Holter monitors in patients with syncope or presyncope. Am J Med 2003;115:1 – 5.
13. Hoch JS, Rockx MA, Krahn AD. Using the net benefit regression framework to con-struct cost-effectiveness acceptability curves: an example using data from a trial of external loop recorders versus Holter monitoring for ambulatory monitoring of ‘community acquired’ syncope. BMC Health Serv Res 2006;6:68.
14. Zimetbaum PJ, Kim KY, Josephson ME, Goldberger AL, Cohen DJ. Diagnostic yield and optimal duration of continuous-loop event monitoring for the diagnosis of pal-pitations. A cost-effectiveness analysis. Ann Intern Med 1998;128:890 – 5. 15. Locati ET, Vecchi AM, Vargiu S, Cattafi G, Lunati M. Role of extended external loop
recorders for the diagnosis of unexplained syncope, pre-syncope, and sustained palpitations. Europace 2014;16:914 – 22.
16. Flint AC, Banki NM, Ren X, Rao VA, Go AS. Detection of paroxysmal atrial fibril-lation by 30-day event monitoring in cryptogenic ischemic stroke: the Stroke and Monitoring for PAF in Real Time (SMART) Registry. Stroke 2012;43:2788 – 90. 17. Mu¨ller A, Scharner W, Borchardt T, Och W, Korb H. Reliability of an external loop
recorder for automatic recognition and trans-telephonic ECG transmission of atrial fibrillation. J Telemed Telecare 2009;15:391 – 6.
18. Linzer M, Pritcett EL, Pontinen M, Mc Carthy E, Devine GW. Incremental diagnostic yield of loop electrocardiographic recorders in unexplained syncope. Am J Cardiol 1990;66:214 – 9.
19. Gula LJ, Krahn AD, Massel D, Skanes A, Yee R, Klein GJ. External loop recorders: determinants of diagnostic yield in patients with syncope. Am Heart J 2004;147: 644 – 8.
20. Brignole M, Moya A, Menozzi C, Garcia-Civera R, Sutton R. Proposed electrocar-diographic classification of spontaneous syncope documented by an implantable loop recorder. Europace 2005;7:14 – 8.
21. Krahn AD, Klein GJ, Yee R, Hoch JS, Skanes AC, Barrett PM et al. Comparison of 24-hour Holter monitoring with 14-day novel adhesive patch electrocardiographic monitoring. Am J Med 2014;127:95.e11 – 7.
22. Giada F, Gulizia M, Francese M, Croci F, Santangelo L, Santomauro M et al. Recur-rent unexplained palpitations (RUP) study comparison of implantable loop record-er vrecord-ersus conventional diagnostic strategy. J Am Coll Cardiol 2007;49:1951 – 6. 23. Park MH, de Asmundis C, Chierchia GB, Sarkozy A, Benatar A, Brugada P. First
ex-perience of monitoring with cardiac event recorder electrocardiography Omron system in childhood population for sporadic, potentially arrhythmia-related symp-toms. Europace 2011;13:1335 – 9.
24. Iglesias JF, Graf D, Forclaz A, Schlaepfer J, Fromer M, Pruvot E. Stepwise evaluation of unexplained syncope in a large ambulatory population. Pacing Clin Electrophysiol 2009;32:S202 – 6.
25. Davis S, Westby M, Pitcher D, Petkar S. Implantable loop recorders are cost-effective when used to investigate transient loss of consciousness which is either suspected to be arrhythmic or remains unexplained. Europace 2012;14:402 – 9.